US5877377A

US5877377A - Metal oxide catalyst and use thereof in chemical reactions

Info

Publication number: US5877377A
Application number: US08/546,305
Authority: US
Inventors: Stanislaw E. Golunski; John M. Gascoyne; Anthony Fulford; John W. Jenkins
Original assignee: Johnson Matthey PLC
Current assignee: Johnson Matthey PLC
Priority date: 1993-08-14
Filing date: 1995-10-20
Publication date: 1999-03-02
Anticipated expiration: 2013-12-10

Abstract

A method for conducting a chemical reaction with a catalyst composed of metal oxide particles among which are uniformly incorporated, in order to reduce the operating temperature of the catalyst, palladium particles.

Description

CROSS REFERENCE TO RELATED APPLICATION

This application is a division of Ser. No. 08/164,413, filed Dec. 10, 1993, now U.S. Pat. No. 5,478,528; and a continuation-in-part of Ser. No. 08/283,569, filed Aug. 1, 1994, now U.S. Pat. No. 5,593,935.

This invention relates to a metal oxide catalyst, its production, a method of conducting a chemical reaction employing it, and a hazardous gas sensor containing a particular example of it.

It has been discovered how a metal oxide catalyst can be operated at a lower temperature.

The invention provides a catalyst composed of metal oxide particles among which are uniformly incorporated, in order to reduce the operating temperature of the catalyst, palladium particles.

The invention provides also a process for preparing the catalyst, which process comprises co-precipitating the metal oxide particles and the palladium particles.

The invention also provides a method of conducting a chemical reaction employing the catalyst.

The invention provides also a sensor of hazardous gas at ambient temperature, which sensor comprises means to allow gas to contact the present catalyst wherein the metal oxide comprises iron(III) oxide and means to indicate the rise in temperature of the catalyst if hazardous gas is present.

The invention provides, in a catalyst composed of metal oxide particles, the improvement which comprises incorporating palladium particles uniformly among the metal oxide particles to reduce the operating temperature of the catalyst.

The invention similarly provides, in a method of conducting a chemical reaction employing a catalyst composed of metal oxide particles, the improvement comprising incorporating palladium particles uniformly among the metal oxide particles to reduce the temperature of the reaction.

It has been discovered that the operating temperature of a catalyst composed of metal oxide particles can be reduced by incorporating palladium particles uniformly among them. In this structure, there is high and even interaction between the two sets of particles. This contrasts, for instance, with metal oxide particles whose surface has simply been impregnated with the palladium.

As is conventional in this art, references herein to a catalyst component being palladium embrace the possibility of some or all of it being in the form of the oxide.

The present catalyst usually contains 0.1 to 25%, preferably 1 to 20%, by mass of the palladium particles based on the total mass of the palladium particles and the metal oxide particles.

Additives can be incorporated into the catalyst to convey advantageous properties or avoid disadvantageous properties. The additives can be conventional. The additives can be for instance antimony oxide, or alkali metal ions to improve selectivity in partial oxidation reactions. Additives can be present for instance in amounts of 0.1-50% of the total mass of the catalyst. The lower amounts are appropriate for additives such as alkali metal ions, and the higher for additives such as antimony oxide. The usual antimony oxide can be employed, generally antimony(V) oxide or that known as diantimony tetroxide.

The lowering of the operating temperature of catalysts composed of metal oxide particles is a general phenomenon (providing, of course, thermodynamic considerations do not render this impossible). It is of particular interest where the metal oxide comprises reducible metal oxide, ie the metal oxide is capable of reduction to another oxide of the metal.

The metal oxide may be any which is known to be catalytically active. It can be the oxide of a single metal (for instance iron(III) oxide, cerium(IV) oxide, niobia, magnesium oxide, vanadia or antimony oxide) or a mixture thereof (for instance a mixture of cerium(IV) oxide and antimony oxide or a mixture of vanadia and magnesium oxide), a mixed metal oxide (for instance bismuth molybdate) or a mixture thereof, or a solid solution of one metal oxide in another (which is not necessarily stoichiometric) or a mixture thereof, or a mixture of more than one of these types. The usual antimony oxide can be employed, generally antimony(V) oxide or that known as diantimony tetroxide.

The particle diameter of the catalyst, as measured by sieving, is usually less than 150 microns.

Preferably the catalyst is such that in one or more of the following reactions, it reduces the temperature at which 9% mol conversion occurs by at least 50° C., preferably at least 100° C., compared to that in the case of the catalyst without the palladium:

(A) the conversion of but-1-ene to butadiene using a gas mixture of but-1-ene and air (1/6 by volume) at a flow-rate of 100 cm^-3 min^-1 per g of the catalyst;

(B) the conversion of carbon monoxide to carbon dioxide using a gas mixture of carbon monoxide and air (1/99 by volume) at a flow-rate of 2000 cm³ min^-1 per g of the catalyst;

(C) the conversion of carbon monoxide and steam to carbon dioxide and hydrogen using a gas mixture of by volume 0.1% carbon monoxide, 10% steam and the balance nitrogen, at a flow-rate of 1250 cm³ min^-1 per g of the catalyst; and

(D) the conversion of isobutane to isobutene using a gas mixture of isobutane and air (1/2 by volume) at a flow-rate of 100 cm³ min^-1 per g of the catalyst.

The catalyst is preferably that preparable by co-precipitation of the metal oxide particles and the palladium particles. Co-precipitation is a very effective method of incorporating the palladium particles into the metal oxide particles, but an alternative preparation which gives the same result would suffice. The co-precipitation can be carried out in a manner known generally per se, conveniently at ambient temperature. The co-precipitation is preferably carried out so that it occurs in a controlled rather than a sudden manner.

It will be understood that the co-pecipitation may produce a precipitate which does not have the metal oxide present as such, but in a form, such as an hydroxide, which is then converted to the metal oxide. The conversion can be accomplished for instance by heating, for example at 50°-500° C.

The chemical reaction in which the present catalyst is employed can be any in which a catalyst without the palladium can be employed. Where the unmodified metal oxide acts as a catalyst in several different reactions, each reaction may be made to occur at a lower operating temperature by means of the present invention. The present reaction is usually conducted at a temperature below 700° C., usually at a temperature within the range of ambient temperature up to 700° C.

By being able to operate a reaction at a lower temperature, a saving of energy can be achieved and the catalyst will tend to last longer. In addition, reactions can now be conducted at ambient temperature which previously required heating the catalyst. In a particular embodiment, the present reaction is conducted at ambient temperature. The low temperature activity of the present catalyst is sustained rather than being transitory. It usually lasts, without regeneration, for at least 5 hours, preferably at least 100 hours, particularly when the catalyst is prepared by co-precipitation.

It is another advantage of the present invention that the catalyst can be employed without prior calcination to activate it. It can simply be washed and dried (at no more than 130° C.) and then used for low temperature catalysis. However, calcination may be desirable to ensure physical stability.

The present chemical reaction is usually oxidation. A preferred reaction is oxidative dehydrogenation, particularly of alkene, especially of but-1-ene to butadiene.

A preferred metal oxide comprises (i.e. consists of or includes) bismuth molybdate. Another preferred metal oxide comprises iron(III) oxide. A further preferred metal oxide comprises cerium(IV) oxide optionally in admixture with antimony oxide. Catalysts containing these metal oxides can catalyse the reactions normally associated with the unmodified metal oxide from which they are derived.

The catalyst wherein the metal oxide comprises iron(III) oxide is able to oxidise carbon monoxide to carbon dioxide, even at ambient temperature. It is not deactivated by the presence of water vapour. Hence, in an advantageous embodiment it is employed to oxidise carbon monoxide to carbon dioxide in the presence of 0 to 15% water vapour. Gas mixtures referred to in this specification are by volume unless otherwise indicated. This catalyst will also tolerate gas mixtures containing nitrogen oxides and/or sulphur compounds. Hence, in advantageous embodiments it is employed to oxidise carbon monoxide to carbon dioxide in the presence of 0 to 0.2% nitrogen oxides and/or 0 to 0.005% sulphur compounds. It has been found that the rate of CO conversion by this catalyst is linearly dependent on CO concentration, over a range of up to 5% or more by volume in gas such as air. The energy released during reaction is, therefore, proportional to the CO concentration in the gas, making this catalyst particularly suitable to be used for CO sensing. It may also be used to sense, in the absence of CO, other hazardous gases, usually a reducing gas, such as hydrogen or alkene, for instance but-1-ene. In a preferred embodiment, the sensing is at ambient temperature, hence without the need for pellister technology.

The sensor can be of type known in itself. Usually the sensor comprises means to allow gas to contact the catalyst wherein the metal oxide comprises iron(III) oxide and means to indicate the rise in temperature of the catalyst if hazardous gas, especially CO, is present. A particular advantage of sensing at ambient temperature is that the catalyst does not have to be kept at a raised temperature, so avoiding the risk of igniting combustible gas. The present catalyst wherein the metal oxide comprises iron(III) oxide is useful for monitoring the performance of catalytic material for oxidising carbon monoxide to carbon dioxide, for instance in an engine exhaust. Accordingly, in a preferred embodiment, this catalyst is downstream of catalytic material in the exhaust system of an engine, the catalytic material being for oxidising carbon monoxide to carbon dioxide and the sensor monitoring the performance of the catalytic material in this oxidation. This catalyst is particularly suited to automobile applications, and can there be used in on-board diagnostics, such as monitoring the performance of catalytic material for treating the automotive exhaust to combat air pollution. Hence, in a preferred embodiment, the engine is an internal combustion engine in a vehicle and the monitoring indicates when the performance of the catalytic material (for instance a three-way catalyst) falls below a set level. Compared to a prior art low temperature CO oxidation catalyst (Au/Fe₂ O₃, see page 33 of "Successful Design of Catalysts", edited by T. Inui, published by Elsevier, Amsterdam, 1988), the present catalyst wherein the metal oxide compises iron(III) oxide has the advantages of (a) lower material costs, (b) lower light-off temperature at the same loading of precious metals, and (c) greater resistance to deactivation.

In the presence of CO and H₂ O, the present catalyst wherein the metal oxide comprises iron(III) oxide functions as a water-gas shift catalyst. This activity begins about 100° C. The catalyst activates in situ, so pre-reduction is not necessary. It can be used in the water-gas shift reaction at, for instance, 100°-200° C.

The present catalysts, particularly those wherein the metal oxide comprises iron(III) oxide or bismuth molybdate, are active in the oxidative dehydrogenation of alkene. The alkene is usually acyclic, and can be straight or branched chain. It generally is of 2-6 carbon atoms. The oxidative dehydrogenation of but-1-ene to butadiene is of particular importance, and the present catalyst effects this at much lower temperatures than expected. This reaction with the present catalyst wherein the metal oxide comprises iron(III) oxide will begin even if the initial temperature of the reactor is below 100° C.; once the reaction has started, for instance after several minutes, it can become self-sustaining without the further supply of heat. In a preferred embodiment, the oxidative dehydrogenation of but-1-ene to butadiene is conducted employing as catalyst the present catalyst wherein the metal oxide comprises iron(III) oxide at a temperature below 200° C., for instance at a temperature between 80 ° and 200° C.

When the present catalyst wherein the metal oxide comprises iron(III) oxide or bismuth molybdate is employed in the oxidative dehydrogenation of but-1-ene to butadiene, the selectivity to butadiene improves with time on line.

The operating temperature of the present catalyst wherein the metal oxide comprises iron(III) oxide for (a) CO oxidation, (b) water-gas shift, and (c) oxidative dehydrogenation is about 250° C. below that for the corresponding catalyst without the palladium. Similarly, other of the present catalysts, for instance that wherein the metal oxide comprises bismuth molybdate, have a substantially lower minimum operating temperature than the corresponding catalyst without the palladium.

When the metal oxide comprises bismuth molybdate, the oxidative dehydrogenation of but-1-ene to butadiene is preferably conducted at a temperature of 200° to 300° C. This is some 100°-150° C. lower than conventional selective oxidation catalysts require.

The present catalysts, particularly those wherein the metal oxide comprises cerium(IV) oxide, optionally with antimony tetroxide incorporated by mixing into the catalyst, are active in the dehydrogenation, oxidative or not, of alkane of at least 2 carbon atoms to alkene. The alkane is usually acyclic, and can be straight or branched chain. It generally is of 2-6 carbon atoms. A preferred such reaction is the dehydrogenation of isobutane to isobutene.

When the metal oxide comprises cerium(IV) oxide, and as additive diantimony tetroxide is incorporated by mixing into the catalyst, a given yield in the dehydrogenation of isobutane to isobutene occurs about 100° C. lower than in the case of the corresponding catalyst without the palladium. Isobutene is useful for instance for manufacturing the petrol additive methyl t-butyl ether.

Analysis indicates that the present catalyst is usually at least partially amorphous, with metal ions and Pd²⁺ ions on the surface. Chemical analysis of the present dry precipitate wherein the metal oxide is iron(III) oxide shows the presence of Fe³⁺ and Pd²⁺ ions on the surface of a predominantly amorphous bulk. A high degree of interaction between the Pd and Fe phases is inferred from temperature-programmed reduction, which shows substantial shifts (to lower temperatures) of the peaks associated with the reduction of Fe³⁺. Chemical analysis of the present dry precipitate wherein the metal oxide is bismuth molybdate shows a mixture of α-bismuth molybdate (Bi₂ Mo₃ O₁₂ ; monoclinic) and β-bismuth molybdate (Bi₂ Mo₂ O₉ ; monoclinic), with there being no evidence of a crystalline Pd-phase. A conventional Bi--Mo--O catalyst (used for comparison in Comparative Example 2 hereafter) is also a mixture of two bismuth molybdates, but these are the β and γ (Bi₂ MoO₆ ; orthorhombic) allotropes. Surface analysis of the present dry precipitate wherein the metal oxide is cerium(IV) oxide shows that essentially all the palladium is present as Pd²⁺, and all the cerium as Ce⁴⁺.

The present catalyst can be employed as the sole catalyst or together with another catalyst, usually comprising one or more of Pt, Pd, Rh and base metal oxide. The present catalyst can be formulated in the usual way to catalyse chemical reactions. When it is employed as sole catalyst, it generally does not need to be dispersed on a separate high surface area carrier. When it is employed together with another catalyst, a high surface area carrier is often useful to carry both catalysts. For instance, the present catalyst can be dispersed on the carrier and, either before or usually afterwards, the other catalyst can be dispersed on the carrier, for instance in the usual way by impregnating with a precursor and calcining to convert the precursor to the other catalyst. The present catalyst itself preferably has a Brunauer Emmett Teller surface area of at least 50, especially at least 100, m² g^-1. The catalyst can be employed in the form of pellets. It can be employed on a support, preferably a monolith, for instance a honeycomb monolith. The monolith can be metal, in which case it can readily be heated, for instance by passing an electrical current through the metal. Alternatively the monolith can be ceramic. A separate heater of gas can be utilised up-stream of the catalyst to heat gas to facilitate its reaction on the catalyst.

BRIEF DESCRIPTION OF THE DRAWING

FIG. 1 of is a graph showing the effect of varying the CO concentration on CO conversion and the size of the exotherm generated.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The invention is illustrated by the following Examples.

EXAMPLE 1

A Pd--Fe--O catalyst with a nominal Pd loading of 20% (by mass) was prepared using crystalline Fe(NO₃)₃.9H₂ O (52.2 g), aqueous Pd(NO₃)₃ (28.0 g of solution≡2.30 g Pd) and solid Na₂ CO₃ (30.1 g). The two nitrates were added to demineralised water (1 dm³) in a large beaker (2 dm³) fitted with a pH probe and a mechanical stirrer. After dissolving the Na₂ CO₃ (in 250 cm³ demineralised water), the aqueous precipitant was added to the stirred Pd/Fe solution at a rate of 2 cm³ min^-1 (using a peristaltic pump).

The Pd/Fe solution was initially orange and had a pH of 1.3, but turned darker as the precipitant was added. The precipitate began to form at a pH of 2.5, and was accompanied by the evolution of carbon dioxide. As the end-point was approached, the suspension became very dark and viscous, and the pH changed more rapidly. At this point, the rate of addition of the precipitant was decreased (to 1 cm³ min^-1), and then stopped when a pH of 8.5 was reached. The brown gelatinous precipitate was isolated (by filtration), washed thoroughly and dried (110° C.; 16 h). The dry material (12 g) was found to contain 19% Pd (and <0.01% Na) by mass.

A sample (0.20 g) of the dry material (sieved to a particle diameter <150 μm) was tested under a gas mixture of CO/air at a flow rate of 400 cm³ min^-1. Gas mixtures in this specification are by volume unless otherwise indicated. The initial temperature of the catalyst bed was 50° C. FIG. 1 of the accompanying drawings shows the effect of varying the CO concentration on:

(i) CO conversion (the amount of CO converted as % of total gas flow),

(ii) the size of the exotherm generated.

(i) shows a linear dependence on CO concentration (over the range 0-5%), indicating that Pd--Fe--O could be used for CO-sensing. (ii) shows a particularly convenient way of doing this (over the range 0-4%).

EXAMPLE 2

Several Pd--Fe--O catalysts with different Pd loadings were prepared by the controlled addition of aqueous Na₂ CO₃ to a mixed solution of iron(III) nitrate and sodium tetrachloropalladite; the addition of the precipitant was stopped when the pH reached 8.5. An analogous series of prior art Au/Fe₂ O₃ (page 33 of "Successful Design of Catalysts", edited by T. Inui, published by Elsevier, Amsterdamn, 1988) catalysts was prepared by substituting tetrachloroauric acid for the Pd-precursor.

The dry materials were tested under a CO/air (1/99) gas mixture at a gas hourly space velocity of 33000 h^-1. The conversion of CO was measured as a function of gas inlet temperature. Values for T₅₀ (temperature at which CO conversion reaches 50%) were recorded and averaged for several temperature programmed tests. For each loading, the T₅₀ value for the Pd-containing catalyst was lower than for the Au-containing analogue (Table 1).

              TABLE 1
______________________________________
Light-off (T.sub.50) temperatures for the oxidation of CO over Pd--Fe--O
and Au/Fe.sub.2 O.sub.3.
Precious Metal  T.sub.50 /°C.
loading/mass %  Pd--Fe--O Au/Fe.sub.2 O.sub.3
______________________________________
0               275       275
1               75        230
2               66        120
4               23         86
8               60         81
______________________________________

When the best Au-containing catalyst (8% Au/Fe₂ O₃) was exposed to the gas mixture at 140° C., the initial CO conversion was 100%, but declined to <20% over a period of 140 hours. The Pd--Fe--O catalysts did not show the same deactivation. Their activity remained at 100% during the first 10 hours, and was still >80% after 140 hours.

EXAMPLE 3

The ability of 20% Pd--Fe--O (nominal composition; prepared as in Example 1) to convert the CO in an exhaust gas was tested at low temperature (100° C.). A sample (1 g) of the dry material was exposed to a simulated mixture of automotive exhaust gas, at an equivalence ratio (λ) of 0.98 (Table 2) and flow rate of 2 dm³ min^-1. The conversion of CO was 95%, and showed no signs of declining with repeated testing (amounting to a total of 20 hours' use).

              TABLE 2
______________________________________
Composition of simulated exhaust gas mixture at λ = 0.98.
Component    Concentration/mol %
______________________________________
H.sub.2      0.43
O.sub.2      0.72
CO           1.30
CH.sub.4     0.067
C.sub.3 H.sub.8
             0.0223
C.sub.3 H.sub.6
             0.0223
CO.sub.2     15.0
H.sub.2 O    9.2
NO           0.15
SO.sub.2     0.002
N.sub.2      balance
______________________________________

EXAMPLE 4

Samples of 20% Pd--Fe--O (prepared as in Example 1) were aged under a variety of conditions before being re-tested under the exhaust gas mixture (at λ=0.98; Table 2). The ageing conditions and the activity results are summarised in Table 3.

              TABLE 3
______________________________________
Effect of catalyst ageing on CO activity under an exhaust gas at
100° C.
Ageing conditions
                Temper-  Duration,
                                H.sub.2
                                      CO conversion
SO.sub.2
        H.sub.2 O
                ature, °C.
                         hour   or O.sub.2
                                      % at λ = 0.98
______________________________________
(a) 0.002%  10%     400    0.5    1% H.sub.2
                                         0
(b) 0.002%  10%     200    5.0    1% O.sub.2
                                        82
(c) 0.002%  absent  400    0.5    1% H.sub.2
                                        55
(d) absent  10%     200    0.5    1% H.sub.2
                                        86
(e) 0.002%  absent  200    5.0    1% H.sub.2
                                        35
(f) absent  absent  400    5.0    1% H.sub.2
                                         0
(g) absent  10%     400    5.0    1% O.sub.2
                                        75
(h) absent  absent  200    0.5    1% O.sub.2
                                        93
______________________________________

Complete loss of low-temperature activity occurred only when the material was aged under a reducing gas at high temperature (ie samples (a) and (f) in Table 3).

EXAMPLE 5

The ability of 20% Pd--Fe--O (prepared as in Example 1) to catalyse the water-gas shift reaction at low CO concentration (in the presence of a large excess of H₂ O) was measured over the temperature range 100°-200° C. A sample (2 g) was tested in a spinning-basket reactor (2500 rpm), using a gas mixture containing CO/H₂ O (1/100) in nitrogen at a flow rate of 2.5 dm³ min^-1. The results are shown in Table 4.

              TABLE 4
______________________________________
Steady-state activity data for water-gas shift reaction.
(Inlet and outlet concentrations/mol ppm)
          Temperature
Catalyst  °C.
                    Inlet (CO)
                             Outlet (CO)
                                     Outlet (H.sub.2)
______________________________________
20% Pd--Fe--O
          100       998      830     112
          110       1002     818     154
          120       998      787     186
          150       998      706     251
          200       1000     650     340
α-Fe.sub.2 O.sub.3
          100       1010     998      0
          150       999      995      0
          250       998      979      0
Fe--O*    250       1006     978      0
______________________________________
 *mildly reduced α-Fe.sub.2 O.sub.3 (0.5% H.sub.2 ; 360° C.;
 20 min)

The difference between the rate of CO conversion and the rate of H₂ release, at temperatures between 100° and 150° C., suggests that 20% Pd--Fe--O was being reduced by the hydrogen being generated. At 200° C., this in situ reduction appeared complete, and the two rates became almost identical. The commercial sample of α-Fe₂ O₃ showed negligible H₂ formation under identical conditions, even after mild reduction.

EXAMPLE 6

In order to determine the extent to which water-gas shift can contribute when 20% Pd--Fe--O (prepared as in Example 1) is exposed to an exhaust gas, the CO conversion was measured both in the presence and absence of O₂ in the simulated exhaust gas (see Table 2 for composition). At 100° C., the CO conversion dropped substantially (Table 5) when O₂ was removed from the simulated exhaust gas; the effect was much less at 165° C. The results show that direct oxidation of CO occurs at the lowest temperatures, but the water-gas shift reaction begins to predominate above 150° C.

              TABLE 5
______________________________________
Effect of O.sub.2 removal from exhaust gas on CO-conversion
over 20% Pd--Fe--O.
          CO-conversion/%
Temperature °C.
            Full exhaust gas
                        Exhaust gas without O.sub.2
______________________________________
100         95          32
140         99          43
165         99          77
______________________________________

EXAMPLE 7

4% Pd--Fe--O (nominal composition) was prepared as described in Example 1, except the mass of Pd in the nitrate precursor was 0.46 g. A sample (1 g) of the dry precipitate was tested under a gas mixture of but-1-ene/air (1/6) at a flow-rate of 100 cm³ min^-1. Unlike the performance expected for catalysts derived from iron oxide (eg see Zhang et al, J. Chem. Soc. Faraday Trans., 88 (1992) 637), Pd--Fe--O showed substantial activity (including oxidative dehydrogenation) at temperatures below 200° C. (Table 6).

              TABLE 6
______________________________________
Butene oxidation activity of fresh 4% Pd--Fe--O as a function of furnace
temperature; activity measured after 20 min on line.
                Selectivity %
Temperature               Trans  Cis
°C.
        Conversion %
                   CO.sub.2
                          but-2-ene
                                 but-2-ene
                                        butadiene
______________________________________
180     74         30.5   30     22     4.5
135     61         43     23     18     15
 80     44.5       45     17.5   15     22
(molar conversions and selectivities)
______________________________________

At a furnace temperature of 80° C., the catalyst bed temperature rose to ca 130° C. during reaction. The heat generated was then sufficient to sustain the reaction without further heat input from the furnace.

The selectivity of Pd--Fe--O to butadiene improved as a function of time on line (Table 7), and could be further enhanced by adjusting the butene/air ratio in the gas feed (Table 7).

              TABLE 7
______________________________________
Enhancing oxidative dehydrogenation of butene over
4% Pd--Fe--O at 100° C.
(a) Effect of conditioning
(activity measured for butene/air = 1/6)
                Selectivity
         Conversion       Trans  Cis
         %         CO.sub.2
                          but-1-ene
                                 but-2-ene
                                        butadiene
______________________________________
2 min on line
         65        45     14     12     15
5 h on line
         54        36      3      2     59
* 2 min on line
         70        42     13     11     21
______________________________________
* after "regeneration" of the 5 h used catalyst under air at 500°
C.
(b) Effect of gas composition
(activity of sample aged on line for 5 h)
                Selectivity %
                          Trans  Cis
Butene/Air
        Conversion %
                   CO.sub.2
                          but-2-ene
                                 but-2-ene
                                        butadiene
______________________________________
1/6     54         36     3      2      59
1/4     51         28     2      1      69
1/2     23         31     4      5      60
______________________________________

Comparative Example 1

In order to assess the significance of the results presented in Example 7, a number of related materials were prepared and tested:

(i) Impregnated 4% Pd--Fe--O was prepared by adding aqueous Pd(NO₃)₃ (containing 0.153 g Pd) to FeO(OH) (3.73 g) to form a thick paste, which was heated gently on a hot plate. The warm paste was transferred to an oven (110° C.) for drying (16 h), before calcination (500° C.; air; 2 h).

(ii) Precipitated Pd--O was prepared by adding aqueous Na₂ CO₃ to aqueous Pd(NO₃)₃ (containing 2.30 g Pd). The precipitate was isolated, washed and dried (110° C.; 16 h).

(iii) Precipitated Fe--O was prepared by adding aqueous Na₂ CO₃ to aqueous Fe(NO₃)₃.9H₂ O (52.2 g). The precipitate was isolated, washed and dried (110° C.; 16 h).

(iv) Precipitated 4% Au/Fe₂ O₃ was prepared by the method described in Example 2.

A sample (1 g) of each material was tested under a gas feed of but-1-ene/air (1/6) at a flow rate of 100cm³ min^-1 (Table 8). None of these materials was active at temperatures below 150° C. Between 150° and 200° C., precipitated Pd--O showed reasonable activity, but the selectivity to butadiene was very low. Over a similar temperature range, impregnated 4% Pd--Fe--O was more selective, but the activity was poor; 4% Au/Fe₂ O₃ showed high initial activity, but this was not sustainable. Precipitated Fe--O needed to be used above 300° C. before the yield of butadiene was comparable to that of precipitated Pd--Fe--O at 80°-100° C.

              TABLE 8
______________________________________
Butene oxidation performance at minimum operating temperature
(ie, the minimum temperature at which measurable amounts of product are
formed) (T); activity measured after 20 min on line.
                     Selectivity, %
Catalyst T °C.
                 Conversion, %
                            CO.sub.2
                                 but-2-ene
                                        butadiene
______________________________________
Precipitated
         150     42         54   43     2.5
Pd--O
Precipitated
         300     41         51   14     35
Fe--O
Impregnated
         150      5         30   38     31.5
4% Pd--Fe--O
Precipitated
           150 * 70         38   30     20
Au/Fe.sub.2 O.sub.3
______________________________________
 * rapid deactivation occurs at this temperature

EXAMPLE 8

Pd--Bi--Mo--O, with a nominal Pd-loading of 10% (by mass) and Bi/Mo molar ratio of 2/1, was prepared by co-precipitation. Initially, aqueous Pd(NO₃)₃ (containing 0.45 g Pd) was added to a solution of Bi(NO₃)₃.5H₂ O (6.06 g) dissolved in 30% HNO₃ (20 cm³). Ammonium molybdate (7.86 g dissolved in 10% aqueous ammonia) was then added dropwise, with very rapid stirring; some precipitation occurred during addition. The pH of the resultant suspension was adjusted to 7.4 (using concentrated aqueous ammonia), completing the precipitation of a fine yellow powder. The precipitate was isolated, washed, dried (110° C.; 16 h) and calcined (500° C.; air; 4.5 h).

When a sample (1 g) of this material was tested under but-1-ene/air (1/7) at a flow-rate of 100 cm³ min^-1, the minimum operating temperature was ca 200° C. The yield of butadiene gradually increased over the first few minutes, before stabilising after 20-25 min (Table 9). This activity was at a temperature 150°-200° C. lower than the minimum expected for mixed-metal oxide catalysts (C. F. Cullis et al in "Catalysis", edited by G. C. Bond and G. Webb, Royal Society of Chemistry, London, 1982, page 273).

              TABLE 9
______________________________________
Butene oxidation activity of 10% Pd--Bi--Mo--O at 200° C.,
as a function of time on line.
                   Selectivity, %
Elapsed time, min
          Conversion, %
                     CO.sub.2
                             but-2-ene
                                     butadiene
______________________________________
 2        55         30.5    24.5    45
25        65         23      19.5    57.5
50        60         20      22      58
______________________________________

Comparative Example 2

For comparison (to Example 8), an unmodified bismuth molybdate catalyst (in which the molar ratio of Bi/Mo=2/1) was prepared by a standard route (Ph Batist et al, J. Catal., 25 (1972), 1). Ground Bi(NO₃)₃.5H₂ O (6.06 g) was added to concentrated aqueous ammonia (15 cm³), and stirred for 5 min. The resultant suspension was filtered to isolate the white powder, which was then washed free of ammonia. The powder was added to H₂ MoO₄ (1.03 g) in distilled water (150 cm³), and the mixture was heated under reflux (18 h). The solid product was isolated, dried (110° C.; 2 h) and calcined (500° C.; 2 h).

A sample of the bismuth molybdate (1 g) was tested at 200° C., under but-1-ene/air (1/7) at a flow-rate of 100 cm³ min^-1 (Table 10). Some initial activity was observed, but only for the first few minutes. Sustainable activity was not achieved until the temperature was raised to 350°-400° C., when the material functioned as a very selective catalyst for butadiene formation (Table 10b). On lowering the temperature back down to 200° C., no activity was observed.

              TABLE 10
______________________________________
Butene oxidation activity of unmodified bismuth molybdate, as a
function of time on line.
                   Selectivity, %
Elapsed time, min
          Conversion, %
                     CO.sub.2
                             but-2-ene
                                     butadiene
______________________________________
(a) 200° C.
 2        25         3       61      36
25         3         1       44      55
50         0         --      --      --
(b) 350° C.
 2        75         8       13.5    70
25        83         5.5     14      62.5
45        83         6.5     14      67.5
______________________________________

EXAMPLE 9

Pd/CeO₂, with a nominal Pd-loading of 4% (by mass) was prepared by co-precipitation. A mixed solution was prepared by adding at ambient temperature Ce(NO₃)₃.6H₂ O (126 g of solid) to aqueous Pd(NO₃)₃ (26.04 g of solution≡2.0 g Pd). This solution was added dropwise to a boiling solution of NaOH (37.08 g) dissolved in the minimum amount of distilled water required to dissolve it. The resultant suspension was maintained at 100° C. for 1.25 hours. The precipitate was then isolated (by filtration), washed, dried (110° C.; 16 h) and calcined (700° C.; air; 2 h). Elemental analysis of the material showed it to contain 3.87% Pd (and <0.01% Na) by mass.

When a sample (1 g) was tested under either isobutane/N₂ (1/5) or isobutane/air (1/2) at a flow-rate of 100 cm³ min^-1, optimum activity for the formation of isobutene occurred at 400° C. (Table 11). The Table also shows that the short-term yield of isobutene could be improved by using a physical mixture of the Pd/CeO₂ (0.6 g) and Sb₂ O₄ (0.4 g).

              TABLE 11
______________________________________
Conversion of isobutane at 400° C., as a function of time on
______________________________________
line
(a) Direct dehydrogenation (under isobutane/N.sub.2)
                                  Isobutene
Catalyst Elapsed Time, min
                      Conversion, %
                                  selectivity %
______________________________________
4% Pd/CeO.sub.2
          2           7.5         >98
         20           7           >98
         35           5.5         >98
4% Pd/CeO.sub.2 +
          2           8.5         >98
Sb.sub.2 O.sub.4
         20           5           >98
______________________________________
(b) Oxidative dehydrogenation (under isobutane/air)
                       Selectivity, %
Catalyst Elapsed Time, min
                      Conversion, %
                                 *CO.sub.x
                                      Isobutene
______________________________________
4% Pd/CeO.sub.2
          2           10         61   39
         20           10         58   42
4% Pd/CeO.sub.2 +
          2           15         45   55
Sb.sub.2 O.sub.4
         20            7         80   20
______________________________________
 *CO.sub.x = CO.sub.2 + CO

Comparative Example 3

For comparison (to Example 9), a physical mixture of CeO₂ (0.6 g) and Sb₂ O₄ (0.4 g) was tested under isobutane/N₂ (1/5) and isobutane/air (1/2), at flow-rates of 100 cm³ min^-1 (Table 12). The yields of isobutene were much lower than for either 4% Pd/CeO₂ or 4% Pd/CeO₂ +Sb₂ O₄, with negligible activity under isobutane/N₂ at temperatures below 500° C.

              TABLE 12
______________________________________
Isobutane dehydrogenation activity of CeO.sub.2 + Sb.sub.2 O.sub.4
______________________________________
(a) Direct dehydrogenation (under isobutane/N.sub.2)
Temperature, °C.
             Maximum isobutane conversion, %
______________________________________
400          0
500          1
550          1.5
______________________________________
(b) Oxidative dehydrogenation (under isobutane/air) at 400° C.
                    Selectivity %
Elapsed time, min
            Conversion %  CO.sub.x
                                 Isobutene
______________________________________
 2          7             76     11
20          6             73     15
______________________________________

Claims

We claim:

1. A method of conducting a chemical reaction, which method comprises contacting a reactant with a catalyst composed of metal oxide particles among which are uniformly incorporated, in order to reduce the operating temperature of the catalyst, palladium particles, wherein the reaction is the oxidative dehydrogenation of but- 1-ene to butadiene, and wherein the reaction is conducted at a temperature below 200° C. and the metal oxide comprises iron (III) oxide.

2. A method according to claim 1, wherein the catalyst contains 0.1-25% by mass of the palladium particles based on the total mass of the palladium particles and the metal oxide particles.

3. A method according to claim 1, wherein the catalyst contains antimony oxide as additive.

4. A method according to claim 1 wherein the catalyst is employed without prior calcination to activate it.

5. A method according to claim 1, wherein the reaction is self-sustaining without the further supply of heat once the reaction has started.

6. A method according to claim 1, wherein the catalyst is at least partially amorphous, and has a catalyst surface having metal ions and Pd²⁺ ions on the catalyst surface.

7. A method according to claim 1, wherein the catalyst comprises a catalyst particle comprising metal oxide particles and palladium particles, wherein the palladium particles are uniformly incorporated among the metal oxide particles, in order to reduce the operating temperature of the catalyst.

8. A method according to claim 1, wherein the catalyst particles respectively have a diameter of less than 150 microns.

9. A method according to claim 1, wherein the catalyst is prepared by co-precipitation of the metal oxide particles and the palladium particles.

10. A method of conducting a chemical reaction, which method comprises contacting a reactant with a catalyst composed of metal oxide particles among which are uniformly incorporated, in order to reduce the operating temperature of the catalyst, palladium particles, wherein the reaction is the water-gas shift reaction and the metal oxide comprises iron(III) oxide, and wherein the reaction is conducted at a temperature of 100°-200° C.

11. A method according to claim 10, wherein the catalyst contains 0.1-25% by mass of the palladium particles based on the total mass of the palladium particles and the metal oxide particles.

12. A method according to claim 10, wherein the catalyst contains antimony oxide as additive.

13. A method according to claim 10, wherein the catalyst is employed without prior calcination to activate it.

14. A method according to claim 10, wherein the catalyst is at least partially amorphous, and has a catalyst surface having metal ions and Pd²⁺ ions on the catalyst surface.

15. A method according to claim 10, wherein the catalyst comprises a catalyst particle comprising metal oxide particles and palladium particles, wherein the palladium particles are uniformly incorporated among the metal oxide particles, in order to reduce the operating temperature of the catalyst.

16. A method according to claim 10, wherein the catalyst particles respectively have a diameter of less than 150 microns.

17. A method according to claim 10, wherein the catalyst is prepared by co-precipitation of the metal oxide particles and the palladium particles.